overexpressionoftead-1intransgenicmousestriatedmuscles...

$: OverexpressionofTEAD-1inTransgenicMouseStriatedMuscles ...followedbytheadditionof1%TritonX-100andincubationon ice for 30 min. Membrane fractions were isolated by homoge-nizing tissues$
Overexpression of TEAD-1 in Transgenic Mouse Striated MusclesProduces a Slower Skeletal Muscle Contractile Phenotype*

Received for publication, September 25, 2008, and in revised form, October 28, 2008 Published, JBC Papers in Press, October 31, 2008, DOI 10.1074/jbc.M807461200

Richard W. Tsika‡§1, Christine Schramm‡§, Gretchen Simmer‡§, Daniel P. Fitzsimons¶, Richard L. Moss¶, and Juan Ji‡§

From the ‡Department of Biochemistry, School of Medicine and §Department of Biomedical Sciences, School of VeterinaryMedicine, University of Missouri, Columbia, Missouri 65211 and the ¶Department of Physiology, Medical School, University ofWisconsin, Madison, Wisconsin 53706

TEA domain (TEAD) transcription factors serve importantfunctional roles during embryonic development and in stri-ated muscle gene expression. Our previous work has impli-cated a role for TEAD-1 in the fast-to-slow fiber-type transi-tion in response to mechanical overload. To investigatewhether TEAD-1 is a modulator of slow muscle gene expres-sion in vivo, we developed transgenic mice expressing hemag-glutinin (HA)-tagged TEAD-1 under the control of themuscle creatine kinase promoter.We show that striated mus-cle-restricted HA-TEAD-1 expression induced a transitiontoward a slow muscle contractile protein phenotype, slowershortening velocity (Vmax), and longer contraction and relax-ation times in adult fast twitch extensor digitalis longus mus-cle. Notably, HA-TEAD-1 overexpression resulted in anunexpected activation of GSK-3�/� and decreased nuclear�-catenin and NFATc1/c3 protein. These effects could bereversed in vivo by mechanical overload, which decreasedmuscle creatine kinase-driven TEAD-1 transgene expression,and in cultured satellite cells by TEAD-1-specific small inter-fering RNA. These novel in vivo data support a role forTEAD-1 in modulating slow muscle gene expression.

A functionally important characteristic of mammalian skel-etal muscle is its diversity in fiber type composition, makingeach skeletal muscle uniquely suited for a particular function.This property is brought about by the differential transcriptionof contractile protein gene families during embryonic myogen-esis and by the influence of nerve activity (1, 2). The structure ofskeletal muscle is highly specialized for the generation of forcethat is required for various movement patterns and posturalmaintenance. The force-generating capacity of skeletalmusclesoccurs bymeans of an intricate organization of contractile pro-teins into myofibrils composed of repeating units called sar-comeres, the smallest force-producing unit of a muscle fiber(cell). Of the skeletal muscle contractile proteins, myosin heavychain (MyHC)2 is a motor protein comprising the thick fila-

ment of sarcomeres, and it has been shown to represent amajordeterminant of the unloaded shortening velocity (Vmax) of agivenmuscle ormuscle fiber (3). TheMyHCprotein is encodedby a multigene family that produces four distinct adult-stageMyHC isoforms termed fast type IIb, IIx/d, IIa, and slow type I(or �). Because each MyHC displays distinct biochemical(actin-activated ATPase activity) and physiological (force-ve-locity) properties, their expression pattern has been shown tocontribute to the classification of four primary fiber typesnamed fast IIb, IIx/d, IIa, and slow type I. Therefore, theamount and type of MyHC comprising the sarcomeres of agiven skeletal muscle is of functional importance.A noteworthy feature of adult skeletal muscle is its intrinsic

ability to adapt its fiber-type composition to accommodatefunctional demands imposed by contractile usage patterns. Inthis regard, both human and animal studies have reported per-turbation-specific skeletal muscle adaptations that involvemass, endurance, strength, power, metabolic properties, andMyHC expression pattern/fiber type (4). Recent inquiries intothe molecular basis of skeletal muscle remodeling have pro-vided evidence that stimulation of the calcium-activated cal-cineurin, calmodulin-dependent protein kinase, and proteinkinase D1 (PKD1) signaling pathways cooperate to modulatethe transcriptional activation of slow fiber genes through classII HDAC degradation and the activation of members of theMEF2 and nuclear factor of activatedT-cells (NFAT) transcrip-tion factor families (5–7). In addition to MEF2 and NFAT,other transcriptional regulators, such as MusTRD/GTF3, Six1,Eya1, PGC-1�, PGC-1�, and PPAR�, have been implicated infiber type-specific gene expression (8–13).In our previous work aimed at identifying cis-acting ele-

ment(s) that direct �MyHC slow fiber gene expression andmechanical overload (MOV) responsiveness, we delineated anA/T-rich element (�A/T-rich; �269/�258, 5�-GGAGATAT-TTTT-3�) that appears necessary for �MyHC transgene slowfiber expression and MOV responsiveness (14). By using the�A/T-rich element as bait in a one-hybrid screen of both adultskeletal muscle and MOV-plantaris cDNA libraries, we identi-fied TEAD proteins as the functionally relevant �A/T-richbinding factor. Interestingly, additional experiments revealedspecific TEAD protein binding to a subset of A/T-rich/MEF2

* This work was supported, in whole or in part, by National Institutes ofHealth, NIAMS, Grants AR41464 and AR47197 (to R. T.). The costs of publi-cation of this article were defrayed in part by the payment of page charges.This article must therefore be hereby marked “advertisement” in accord-ance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 To whom correspondence should be addressed: University of Missouri,Columbia, 440D Bond Life Science Center, Columbia, MO 65211. Tel.: 573-884-4547; Fax: 573-884-3087; E-mail: [email protected].

2 The abbreviations used are: MyHC, myosin heavy chain; MOV, mechanicaloverload; TEAD, TEA domain; HA, hemagglutinin; MCK, muscle creatine

kinase; WT, wild type; siRNA, small interfering RNA; EDL, extensor digito-rum longus; Tg, transgenic; RT, reverse transcription; qRT, quantitativereverse transcription; NFAT, nuclear factor of activated T-cells; GSK, glyco-gen synthase kinase.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 283, NO. 52, pp. 36154 –36167, December 26, 2008© 2008 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

36154 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283 • NUMBER 52 • DECEMBER 26, 2008

by guest on July 4, 2020http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

elements located within the control region of several othermuscle-specific genes and revealed that this binding wasenriched duringMOV-induced fast to slow fiber type switching(15).The vertebrate TEAD genes encode a family of transcrip-

tion factors that include TEAD-1 (NTEF-1/TEF-1), TEAD-2(ETEF-1/TEF-4), TEAD-3 (DTEF-1/TEF-5), and TEAD-4(RTEF-1/TEF-3). All TEAD familymembers contain an evolu-tionarily conserved 72-amino acid DNA binding domain(TEAD) that is found in plant (ABAA), fly (Scalloped), and yeast(TEC1) transcription factors. TEAD proteins have been shownto serve a regulatory role by binding to canonical MCAT ele-ments (5�-CATTCC(T/A)-3�) located in the promoter/en-hancer region of several cardiac, smooth, and skeletal musclegenes and by combinatorial interactions with adjacently boundtranscriptional regulators andubiquitous and/or tissue-specificcoactivators (16, 17). Although most mammalian tissuesexpress more than one TEAD protein during embryonic andadult life, several recent studies involving single or doubleTEAD gene inactivation (TEAD-1, TEAD-2, and TEAD-4)have shown that these transcription factors serve both uniqueand overlapping functional roles during embryonic develop-ment (18–22). Thus, the regulatory specificity of each TEADprotein during development and in response to various stimuliis probably due to their capacity to recognize a broad spectrumof DNA elements, accessibility to target elements, interactionswith various tissue-specific co-activator/co-binding proteins,post-translational modifications, or subtle gradients in theirexpression levels.To investigate whether TEAD-1 functions as a modulator of

basal and/or inducible slow muscle gene expression in vivo, wegenerated transgenic mice that expressed HA-tagged TEAD-1under the control of the muscle creatine kinase (MCK) pro-moter. ThemouseMCKpromoter has been extensively studiedin both cultured striated muscle cells and transgenic mice (23–25). In transgenic mice, a 3300-nucleotide region of the MCK5�-flanking sequencewas shown to be sufficient to drive expres-sion of a reporter gene at high levels in skeletal muscle, at lowerlevels in the heart, and at barely detectable levels in nonmuscletissue, which is similar in pattern andmagnitude to the expres-sion pattern of the endogenousMCK gene (23). In addition, wehave previously shown that 48 h of MOV virtually repress theexpression of the 3300-bp MCK promoter/reporter gene (26).On the basis of the latter data, we hypothesize that any pheno-type resulting from MCK-driven TEAD-1 overexpressioncould in part be reversed by 48 h of MOV. We show that apersistent increase in TEAD-1 protein induced a change inMyHC and troponin complex protein expression pattern andcontractile properties that more closely resemble slow oxida-tive muscle fibers. We further demonstrate that increased HA-TEAD-1 expression activated glycogen synthase kinase (GSK)-3�/3�, resulting in decreased nuclear �-catenin and NFATc1/c3. The latter effects could be reversed in vivo by 48 h of MOV,which decreased MCK-driven TEAD-1 transgene expression,and in cultured satellite cells by TEAD-1 siRNA. These datasupport a novel role for TEAD-1 in modulating a slow skeletalmuscle gene program.

EXPERIMENTAL PROCEDURES

Generation and Screening of Transgenic Mice—Transgenicmice were generated by microinjection of purified transgeneDNA into pronuclei of single cell fertilized embryos, asdescribed previously (27). Transgene-positive mice (founders)were identified using PCR. Subsequent offspring derived frommating the founders were also screened for transgene incorpo-ration by using PCR. In the present study,multiple independenttransgenic lines representing the HA-TEAD-1 transgene werestudied, and all linesweremaintained in a heterozygous state bycontinual outbreeding to nontransgenic BDF1mice. The trans-gene was designed as follows. The mouse TEAD-1 cDNA con-taining an in-frame HA tag at its 5�-end was subcloned at the3�-end of the MCK regulatory sequences. The HA-taggedTEAD-1 cDNA was released from pBluescriptSK by XhoIdigest and subcloned into the blunt-ended PstI site of plasmidpBSMCK (a generous gift fromDr. R. Kahn), which consisted ofa 6.5 genomic region composed of 3.3 kb of theMCK promoterand enhancer 1, untranslated exon 1, 3 kb of intron 1, whichincluded the enhancer 2 region, and the first 16 bp of exon 2(28). The TEAD-1 cDNA was followed by the SV40 polyad-enylation signal. The transgene was released from the plasmidby KpnI/SacII digest, gel-purified, and injected into fertilizedeggs of BDF1 mice. The incorporation of an HA tag allowed usto easily make a distinction between endogenous TEAD-1 pro-tein and the transgene-encoded HA-TEAD-1 protein byWest-ern blot analysis (Fig. 1A).Mechanical Measurements—Force-velocity and power-load

relationships on adult (5-month-old) sex-matched wild-type(WT) andHA-TEAD-1 transgenicmice (n� 4–6mice/group)were determined using Dynamic Muscle Control software(Aurora Scientific) to elicit tetanic afterloaded contractions aspreviously described (29). Contractile measurements werecompleted with the assessment of maximal isometric contrac-tion (Po). Following the completion of mechanical measure-ments, the extensor digitorum longus (EDL) muscle wasremoved from the experimental apparatus, blotted on filterpaper, and immediately weighed. The total cross-sectional area(mm�2) for each muscle was determined by dividing musclemass (mg) by the product of optimal fiber length and 1.06 mgmm�3, the density of mammalian skeletal muscle (30). Force-velocity and power-load curves were analyzed as described pre-viously (31). The values for twitch and tetanic tension werenormalized to cross-sectional area to obtain specific tension(millinewtons mm�2). In a similar fashion, power output(watts) was normalized with respect to muscle mass (wattskg�1) (30).Protein Isolation for Western Blots—For total protein ex-

tracts, 50mg of tissue was homogenized for 30 s on ice in 0.5mlof buffer (50 mM Tris, pH 7.5, 10 mM EGTA, 5 mM EDTA,phenylmethylsulfonyl fluoride, phosphatase inhibitor mixture1, 2, and proteinase inhibitor mixture (Sigma)), using an OmniInternational TH homogenizer. For isolation of nuclear pro-tein, the homogenate was centrifuged at 14,000 � g for 10 minat 4 °C, the cytoplasmic fraction (supernatant) was removed,and the pellet was resuspended in 0.2 ml of buffer. Sonicationfor 10 s with a dismembranator (Fisher) of homogenates was

TEAD-1 Participates in Slow Fiber Gene Expression

DECEMBER 26, 2008 • VOLUME 283 • NUMBER 52 JOURNAL OF BIOLOGICAL CHEMISTRY 36155


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

followed by the addition of 1% Triton X-100 and incubation onice for 30 min. Membrane fractions were isolated by homoge-nizing tissues in 1.5 ml of buffer (50 mM Tris-HCl, pH 7.4, 50mM mannitol, 2 mM EDTA), centrifugation at 500 � g for 10min at 4 °C,mixing the supernatantwith 3.2ml of buffer (50mMTris-HCl, pH 7.4, 300 mMmannitol, 2 mM EDTA), and centrif-ugation at 40,000 rpm for 45 min. Protein concentrations weredetermined using a protein assay kit (Bio-Rad), and extractswere stored at�80 °C. Protein separation and analysis was per-formed as previously described (15). Experimental and stand-ard bands were scanned and quantified usingMulti Gauge soft-ware, and the experimental data were normalized by dividingby the signal of the standard. The antibodies used in this studyare as follows: TEAD-1 (BD Transduction Laboratories), HA(Cell Signaling), Akt, p-Akt, Akt1, and Akt2 (Cell Signaling),GSK-3�/� (Santa Cruz Biotechnology, Inc., Santa Cruz, CA),p-Ser-GSK-3� and p-Ser-GSK-3� (Cell Signaling), glyceralde-hyde-3-phosphate dehydrogenase (Cell Signaling), �-catenin(Cell Signaling), NFATc1 and NFATc3 (Santa Cruz Biotech-nology), Serca2a (Badrilla), Serca1 (Santa Cruz Biotechnology),Myoglobin (Santa Cruz Biotechnology), Troponin I slow (SantaCruz Biotechnology), histone H1 (Santa Cruz Biotechnology),IP90 anti-peptide antibody (Abcam), and anti-rabbit IgG(horseradish peroxidase-linked) and anti-mouse IgG (horse-radish peroxidase-linked) (Cell Signaling).Protein Extraction for High Resolution Gel Electrophoresis—

Protein extraction and high resolution gel electrophoresis forMyHC isoform separation were performed as described previ-ously (32). Briefly, 50 mg of muscle tissue of adult (5-month-old) and aged (11-month-old) wild-type and TEAD-1 Tg micewas processed, and 0.75 �g of total protein was separated on an8% acrylamide gel by 104 V for 24 h at 4 °C. The gel was thenstainedwith SilverSNAP stain (Thermo Scientific) according tothe manufacturer’s recommendations.Animal Care andMOV Surgical Procedure—The imposition

of a mechanical overload (bilateral) on the fast twitch plantarismuscle was accomplished as described by us previously (26)and approved by theAnimalCareCommittee for theUniversityof Missouri-Columbia. The animals were housed in an Associ-ation for the Assessment and Accreditation of Laboratory Ani-mal Care International-accredited animal facility.Satellite Cell Isolation and siRNA Transfection—Freshly iso-

lated satellite cells from adult mice were isolated and culturedas previously described (33). Briefly, muscle was dissected fromthe hind limbs, minced, digested in 400 units/ml collagenasetype I (Worthington), diluted in Ham’s F-12 medium (Invitro-gen), filtered, and collected by centrifugation. Cells were cul-tured on treated plates (Nunc) coated with 0.66% gelatin unlessotherwise noted. After 2 days in culture, 50 nM siRNA mouseTEFD1 (Dharmacon RNA Technologies, ON-TARGETplusSMARTpool) and siCONTROL Non-Targeting #1 were trans-fected using Lipofectamine 2000 (Invitrogen) according to themanufacturer’s protocol, and the protein was extracted after 36 hof incubation and analyzed byWestern blot as described above.RNA Samples and Isolation—EDLmuscles of adult wild-type

and transgenic (RTN12) mice were surgically removed andtransferred to RNAlater (Ambion) according to the manufac-turer’s instructions. Total RNA was isolated from �20–50 mg

of tissue from each animal using RNA STAT-60 according tothe manufacturer’s instructions (Tel Test Inc). Equal amountsof total RNA from each of four replicate tissue samples werepooled for cDNA synthesis after RNA quality control tests.RNA Quality Control—Immediately prior to cDNA synthe-

sis, the purity and concentration of RNA samples were deter-mined from A260/280 readings using a dual beam UV spectro-photometer, and RNA integrity was determined by capillaryelectrophoresis using the RNA 6000 Nano Lab-on-a-Chip kitand the Bioanalyzer 2100 (Agilent Technologies, Santa Clara,CA) as per the manufacturer’s instructions.Reverse Transcription, PCR, andReal TimeQuantitative PCR

Analysis—Single strand cDNAwas synthesized from0.02 to 2.0�g of each total RNA sample using the High Capacity cDNAsynthesis kit (Applied Biosystems), according to the manufac-turer’s instructions. Conventional PCR for TEAD isoforms wasperformed using 20 nmol of the following primers forTEAD-1 (5�-AGAGCCCTGCCGAAAACATGG and 5�-TGG-CTGTCCTGTCTGTATCAT), TEAD-2 (5�-GGAAGGCA-GCGAAGAGGGC and 5�-CTTCACGTCTGGAACATTC-CAT), TEAD-3 (5�-TGGACAAGGGTCTGGACAACG and5�-AACCTTGAGGAGGAGGAGAAG), TEAD-4 (5�-ATTA-CCTCCAACGAGTGGAGC and 5�-CTGGCAAAGCTCCT-TGCCAAA), and tubulin (5�-TCACTGTGCCTGAACT-TACC and 5�-GGAACATAGCCGTAAACTGC), 2.5 mMMgCl2, 200 nM dNTPs, and 1 unit of Taq polymerase per reac-tion at 95 °C for 5min at 95 °C, 65 °C, and 72 °C for 30 s each for26–30 cycles, and at 72 °C for 10 min. The PCR product wasanalyzed on a 1.2% standard agarose gel and stained withethidium bromide. Approximately 20 ng of each cDNA (9 �l)was mixed with 10 �l of TaqMan� Gene Expression 2� PCRMaster Mix (Applied Biosystems) and 1 �l of each indicatedTaqMan� gene expression assay (Applied Biosystems) in 384-well plates and analyzed on the 7900HT Fast Real Time PCRsystem according to the manufacturer’s instructions (AppliedBiosystems). Primary analysis of the acquired signal data wereperformed in SDS 2.3 and RQ Manager 1.2 (Applied Biosys-tems). Outlier reactions were removed after Grubb’s test iden-tification, and differential expression was calculated using the��CT method.

RESULTS

Persistent Expression of Transgenic HA-TEAD-1 Is Restrictedto Striated Muscle and Does Not Induce Skeletal MuscleHypertrophy—Our previous work has implicated a role forTEAD-1 in the fast-to-slow fiber type transition in the fasttwitch plantaris muscle as a result of mechanical overload (14,15). To determine the phenotypic consequences of a persistentincrease in striated muscle TEAD-1 protein expression in vivo,we have generatedmultiple transgenicmouse lines that expresshemagglutinin-tagged TEAD-1 (HA-TEAD-1) cDNA underthe control of the mouse muscle-specific creatine kinase regu-lator sequences (Fig. 1A). A comparison of body weight, nor-malized skeletal muscle weight (muscle weight/tibia length ormuscle weight/body weight), or cross-sectional area (data notshown) did not reveal a difference between age and gender-matched WT and TEAD-1 Tg mice. These data demonstratethat an increase in total TEAD-1 protein (endogenous and




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

exogenous (transgenic)) was not sufficient to directly and/orindirectly induce skeletal muscle hypertrophy or to alter thegrowth rate of the TEAD-1 Tg mice. To determine the tissue-specific and developmental expression pattern of the MCK-HA-TEAD-1 transgene, we performed Western blot analysis

using anti-HA antibody and total protein extracts isolated fromskeletal muscles composed of varying proportions of fast andslow fiber types, nonmuscle tissues, and the heart of adult(5-month-old) WT and TEAD-1 Tg mice. Endogenous MCKgene expression is restricted to striatedmuscle (skeletal muscle

FIGURE 1. Schematic of TEAD-1 transgene and expression pattern of transgenic and endogenous TEAD-1. A hemagglutinin-tagged mouse TEAD-1 cDNAwas subcloned at the 3�-end of a mouse muscle-specific creatine kinase promoter/enhancer cassette followed by SV40 polyadenylation signal (A). Western blotanalysis was performed using 100 �g of total protein extract isolated from the soleus (B), plantaris (C), EDL (D), gastrocnemius (E), and heart (F) of WT and fromthree independent TEAD-1 transgenic mouse lines (lines 4, 12, and 14). Analysis using either anti-HA or anti-TEAD-1 antibody revealed the striated muscle-specific expression pattern of the TEAD-1 transgene (HA-tagged TEAD-1 protein). G, HA-tagged TEAD-1 protein was not detected in nonmuscle tissue (brain,liver, spleen, and kidney) of transgenic line 12. TEAD-1 transgenic line 12 embryonic day 17 (H) and neonatal (day 14 postbirth) (I) leg muscle tissue expressHA-tagged TEAD-1 protein.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

and heart) and is first detected in skeletal muscle at approxi-mately embryonic day 17. In addition, the magnitude of MCKgene expression in adult skeletal muscle follows a fiber type-specific pattern wherein the highest level of expression isdetected in fast type IIb fibers, with successively lower levelsoccurring in the order IIx/d � IIa � I (�) (34, 35). Likewise,extensive study of the mouse MCK promoter has revealed thattransgenic mice harboring a 3300-nucleotide region of MCK5�-flanking sequence displayed an expression pattern thatmimicked that of the endogenousMCKgene (23–25). ByWest-ern blot analysis, HA-TEAD-1 protein was detected only in thetotal protein extracts isolated from adult hearts, as well as adult,embryonic day 17 (fetal), and day 14 postnatal skeletal muscleof all three TEAD-1 Tg lines (4, 12, 14) examined (Fig. 1, B–H)(data not shown). The additional HA band that appears in lanes2 and 3 for transgenic line 4 probably represents a truncatedHA-TEAD-1 product due to degradation or a smaller product,initiation of translation at an alternative site, or truncation ofthe transgene during chromosomal integration at a second site.Moreover,HA-TEAD-1 proteinwas not detected in the cellularextracts obtained from the brain, liver, spleen, or kidney ofTEAD-1 Tg mice (Fig. 1G). Consistent with the broad expres-sion pattern of endogenous TEAD-1 protein, anti-TEAD-1-specific antibody detected endogenous TEAD-1 protein in theheart, skeletal muscle, brain, liver, spleen, and kidney of adultWT and TEAD-1 Tg mice (Fig. 1, B–H). Multiple bands weredetected when using anti-TEAD-1 antibody, because this anti-body recognizes a common epitope (amino acids 86–199) pres-ent in both HA-TEAD-1 and endogenous TEAD-1 protein (Fig.1, B–H). Consistent with the pattern of MCK gene expression,HA-TEAD-1 protein was detected at the highest levels in theEDL and plantaris muscles, which contain a predominance offast type II fibers (�90%), whereas comparatively lower levelswere detected in the soleusmuscle (�60–70% type I fibers) andin the adult heart (Table 1). Importantly, althoughHA-TEAD-1protein was detected in the heart of adult HA-TEAD-1 Tg (line12) mice, a significant deficit in cardiac function was notdetected, as assessed by echocardiography and MRI analysis.3Since all lines displayed a similar striated muscle-restrictedHA-TEAD-1 expression pattern, we have chosen to presentonly our analysis ofHA-TEAD-1Tg line 12. Since adult striatedmuscle has been shown to express multiple TEAD protein iso-forms, we employed RT-PCR and qRT-PCR to assess whetherthemRNA levels encoding other TEAD isoproteins in TEAD-1Tg line 12 mice would be altered due to a persistent increase inTEAD-1 protein.With the exception of TEAD-1mRNA levels,mRNAs encoding TEAD-2, TEAD-3, and TEAD-4 did not dif-

fer from the levels detected inWTmouse EDL muscle (Fig. 2Aand Table 2). Collectively, these data demonstrate that theMCK-driven HA-TEAD-1 transgene mimicked the striatedmuscle-restricted expression pattern of the endogenous MCKgene. Additionally, the persistent increase in total TEAD-1 pro-tein did not result in cardiac dysfunction; nor did it directly orindirectly alter the basal gene expression level of other mem-bers of the TEAD gene family.Mice Overexpressing TEAD-1 Display a Transition toward a

SlowerMuscle Contractile Phenotype—Previouswork by us (15,36) and others (16, 37–40) has provided in vitro evidence thatthe TEAD proteins participate in the regulation of several con-tractile proteins. To determine whether a persistent increase inTEAD-1 expression regulated MyHC gene expression in vivo,we performed a comparative high resolution gel electrophore-sis analysis of myofibrillar protein isolated from either WT orTEAD-1 Tg line 12 mice. A striking observation was that theintensity of silver-stained bands representing the fast type IIx/dMyHCwas basically absent in both fast and slow twitchmusclesof the TEAD-1 Tg mice as compared with those of WT mice.Moreover, the plantaris, EDL, and gastrocnemius muscles ofthe TEAD-1 Tg mice displayed a small but reproducibledecrease in the band intensity of fast type IIbMyHC and a smalland reproducible increase in the band intensity of the fast typeIIa MyHC. Interestingly, only the soleus and gastrocnemiusmuscles of the TEAD-1 Tg line 12 mice showed an increase inthe band intensity of the slow type I (�) MyHC despite a signif-icant increase in type I MyHC transcripts in the EDL muscle(Fig. 3, A–D, and Table 3). These data demonstrate that3 R. W. Tsika, et al. manuscript in preparation.

FIGURE 2. Expression analysis of the TEAD gene family in adult striatedmuscle. RT-PCR analysis of adult EDL muscle RNA (1 �g) isolated from WT andTEAD-1 transgenic line 12 (Tg) EDL muscle showed increased TEAD-1 mRNAlevels without compensatory alteration of TEAD-2, TEAD-3, or TEAD-4 mRNA.

TABLE 1Densitometry quantification of TEAD-1 overexpression in striatedmuscle tissue of transgenic line 12 (n � 6)

TEAD-1 expression versusWTPlantaris 23-fold1EDL 37-fold1Soleus 2.4-fold1Gastrocnemius 17-fold1Heart 3.3-fold1

TABLE 2qRT-PCR analysis of TEAD1– 4 mRNA expression in adult TEAD-1 TgEDL muscle (line 12; n � 3) verified RT-PCR results showing anincrease in HA-tagged TEAD-1 mRNA without a compensatoryalteration in TEAD-2, TEAD-3, or TEAD-4 mRNA abundance

Gene qRT-PCR RegulationTEAD-1�HA-TEAD-1 18-fold IncreasedTEAD-1 1.2-fold No changeTEAD-2 1.0-fold No changeTEAD-3 1.2-fold No changeTEAD-4 1.1-fold No change




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

TEAD-1 overexpression resulted in anMyHC profile reflectiveof a slower phenotype in both fast and slow twitch skeletalmus-cles. Likewise, lines 4 and 14 displayed MyHC expression

patterns that resembled that ofline 12, indicating that the MyHCexpression pattern observed dueto TEAD-1 overexpression was notinfluenced by the transgene integra-tion site. Furthermore, a compara-tive analysis of muscle extractsobtained from line 12 mice at 5 and11 months of age did not reveal anydifferences in MyHC expressionpatterns (Fig. 3E).It is well established that the con-

tractile velocity of a given muscle ormuscle fiber is strongly correlated totheMyHCcompositionof its sarcom-eres (3, 41). To directly ascertainwhether the putative TEAD-1-in-duced shifts inMyHC isoform profileresulted in alterations in the overallpower-generating capabilities of thefast twitch extensor digitorum longusmuscle of adult TEAD-1 Tg line 12mice, we performed a comparativecharacterization of the force-velocityand power-load relationship betweenWT (n � 6) and TEAD-1 Tg (n � 6)EDL muscle. Shortening velocitieswere slower in the TEAD-1 Tg miceas compared with theWTmice at allrelative loads less than isometric (Po;Fig. 3E). This effect resulted in amarked depression in the relativepower output in the EDL muscle ofTEAD-1 Tg mice (0.32 0.02 versus0.51 0.03 P/Po�ml/s; Fig. 3F). Con-comitant with the reduction in rela-tive power output, overexpression ofTEAD-1 elicited a�40% reduction inboth peak absolute power (1.2 0.2versus 2.3 0.4 watts; p 0.05) andnormalized power output (55.9 6.4versus 93.6 10.7 watts kg�1; p 0.05).To further investigate the effect

of TEAD-1 overexpression on thecontractile properties of the fasttwitch EDL muscle, we assessedtwitch kinetics. The contractileproperties of the TEAD-1 Tg line 12and WT EDL muscle are summa-rized in Table 4. The EDLmuscle ofTEAD-1 Tg mice (n � 6) displayedsignificantly prolonged twitch dura-tion (92.0 3.0 versus 79.0 4.0ms) and half-relaxation time

(18.0 1.0 versus 14.0 1.0ms)when comparedwithWTEDL(n � 6) muscle. No significant difference in either twitch ten-sion or tetanic tension betweenWT andTEAD-1 Tgmice were

FIGURE 3. Representative MyHC protein expression pattern in various muscles and TEAD-1-inducedcontractile properties of the fast twitch EDL muscle. High resolution glycerol gel electrophoresis of proteinextracts (0.75 �g) isolated from plantaris (A), EDL (B), soleus (C), and gastrocnemius (D) muscles of WT andTEAD-1 Tg (line 12) mice revealed that a persistent increase in TEAD-1 expression results in a fast-to-slowtransition in skeletal muscle MyHC composition in both fast and slow twitch skeletal muscles. E, plantaris MyHCseparation of WT and Tg lines 12, 4, and 14 as well as 5- and 11-month Tg line 12 show no difference betweenTg lines and age. F, cumulative force-velocity and power-load relationships of EDL muscles isolated from WTand TEAD-1 Tg mice (line 12). Shortening velocities and power output were reduced over most relative loads inthe TEAD-1 Tg EDL muscle, as evidenced by a �40% decrease in peak normalized power.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

observed. The slowing of twitch kinetics is consistent withchanges in theMyHC isoform profile as well as altered kineticsofCa2�handling by the sarcoplasmic reticulumandCa2�bind-ing to the thin filament or a combination of all three (42).To determine whether TEAD-1 overexpression resulted in

the transcription of genes encoding proteins involved in Ca2�

handling by the sarcoplasmic reticulum and Ca2� binding tothe sarcomeric thin filament, we used qRT-PCR to obtain anassessment of transcripts encoding the slow sarcoplasmic retic-ulum Ca2�-ATPase (SERCA2a), the fast sarcoplasmic reticu-lum Ca2�-ATPase (SERCA1), and the slow isoforms of tropo-nin T, troponin I, and troponin C. When compared with theWT EDL, transcripts encoding the slow sarcoplasmic reticu-lum Ca2� ATPase and the slow isoforms of troponin T, tropo-nin I, and troponin C were increased in the TEAD-1 Tg line 12EDL muscle 4.7-, 3.6-, 7.1-, and 6.1-fold, respectively, whereastranscripts encoding fast SERCA1 decreased 1.7-fold (Table 5).Consistent with an increase in mRNA transcripts encodingsTnI and SERCA2a,Western blot analysis using protein extractobtained from the EDL muscle of TEAD-1 Tg mice and anti-sTnI- or anti-SERCA2a-specific antibodies detected higher lev-

els of sTnI and SERCA2a protein. Likewise, Western blot anal-ysis using anti-SERCA1 antibody confirmed decreased fastSERCA1 expression (Fig. 4). In addition, Western blot analysisdetected an increase in myoglobin protein, an oxygen bindingprotein that is highly expressed only in oxidative skeletal mus-cle fibers (Type I � IIa � IIx/d), providing further evidence fora fast-to-slow fiber-type transition in the EDL muscle ofTEAD-1Tgmice (Fig. 4). Together, these results provide strongin vivo evidence suggestive of a regulatory role for TEAD-1 inthe transcription of genes encoding proteins highly representedin slow oxidative myofibers.TEAD-1 Overexpression Decreases Nuclear NFAT Levels—Pre-

viouswork has shown that activation of the calcium-dependentphosphatase calcineurin results in the nuclear translocation ofmembers of the NFAT transcription factor family and theirsubsequent interaction with other transcriptional activators tomediate slow muscle gene expression (6, 43). To determine ifTEAD-1 overexpression altered the levels of nuclear NFATprotein, we performed aWestern blot analysis using both anti-NFATc1- and anti-NFATc3-specific antibodies and nuclearextracts obtained from the EDL and plantaris muscle of WTand TEAD-1 line 12 Tg mice. There was a significant decreasein NFATc1 and NFATc3 protein in the nuclear extracts iso-lated from the EDL and plantaris muscles of TEAD-1 Tg micewhen comparedwith those isolated fromWTmice. In contrast,we did not detect a difference in NFATc1 and NFATc3 proteinbyWestern blot when using total cellular extracts (Fig. 5,A–F).TEAD-1 Overexpression Increases Glycogen Synthase Kinase-3

Activity—In contrast to the phosphatase calcineurin, phospho-rylation of NFAT byGSK-3� has been shown to exclude NFATfrom the nucleus and by this means decreases activation ofNFAT target genes (44, 45). To determine whether TEAD-1overexpression leads to altered GSK-3 activity we performed aWestern blot analysis using anti-GSK3�- or anti-GSK-3�-spe-cific antibody to assess GSK-3 content and activity. No signifi-cant difference in totalGSK-3� orGSK-3�was detected in totalprotein extract obtained from TEAD-1 Tg line 12 EDL or plan-taris muscle when compared with those obtained from WTmice (Fig. 6,A andB). The activity of GSK-3 is tightly regulated,whereby phosphorylation of Ser-21 of GSK-3� and Ser-9 of

FIGURE 4. Contractile properties and expression profile of genes encod-ing proteins highly represented in slow oxidative myofibers. Westernblot analysis using 100 �g of protein extract obtained from adult wild-type(WT) and TEAD-1 Tg (line 12) EDL muscle confirmed qRT-PCR analysis, reveal-ing a fast-to-slow shift via increase in slow troponin I and SERCA2a while fastSERCA1 decreased.

TABLE 3Densitometry quantification of MyHC isoform expression in striatedmuscle tissue of wild-type versus transgenic line 12 (n � 6)

MyHCisoform Plantaris EDL Soleus Gastrocnemius

IIa 1.5-fold1a 1.3-fold1a 1.15-fold1a 1.25-fold1b

IIx 67%2a 81%2c 95%2a 89% fold2b

IIb 1.0 no change 1.0 no change 1.0 no change 10%2b

I NDd ND 1.2 fold1a 1.2 fold1a p 0.005.b p 0.05.c p 0.0005.d ND, not detectable.

TABLE 4Summary of contractile properties shows significantly longercontraction and relaxation times in TEAD-1 Tg EDL muscle (line 12)

Measurement WT Tg, L12Body weight (BW) (mg) 37.6 4.9 31.2 2.0Muscle weight (MW) (mg) 11.1 1.1 10.1 0.4MW/BW (mg g�1) 0.30 0.01 0.33 0.02Lo (mm) 12.5 0.1 12.2 0.1Cross-sectional area (mm2) 1.9 0.2 1.8 0.1Twitch tension (micronewtons mm�2) 28.8 1.6 26.5 2.6Tetanic tension (micronewtons mm�2) 184.2 9.1 203.2 13.9TR50 (ms)a 14 1 18 1bCT (ms)c 79 4 92 3b

a TR50, half-relaxation time.b p 0.05.c CT, twitch duration.

TABLE 5Summary of qRT-PCR (1 �g of RNA) and Western blot (100 �g ofprotein) analysis of adult EDL muscle from wild-type and TEAD-1 Tg(line 12) mice indicates a regulatory role for TEAD-1 in thetranscription of these genes

Gene qRT-PCR RegulationMyHC IIx 5.0-fold DecreasedMyHC IIb 1.4-fold DecreasedMyHC IIa 2.0-fold IncreasedMyHC I (�) 13.7-fold IncreasedTroponin C, slow 6.1-fold IncreasedTroponin I, slow 7.1-fold IncreasedTroponin T1, slow 3.6-fold IncreasedATPase, Ca2�-transporting, slow 4.7-fold IncreasedATPase, Ca2�-transporting, fast 1.7-fold Decreased




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

GSK-3� decreases kinase activity, whereas dephosphorylationof these serine residues activates kinase activity (46, 47). West-ern blot analysis using either phospho-GSK-3� (Ser-21)-spe-cific or phospho-GSK-3� (Ser-9)-specific antibodies revealed asignificant decrease in GSK-3� and GSK-3� phosphorylation(activation) in the protein extracts obtained from the EDL andplantaris muscles of TEAD-1 Tg mice when compared withthose obtained fromWTmice (Fig. 6, A and B).Signaling by Akt family members (Akt1, -2, and -3) has been

shown to inhibit GSK-3�/� activity via phosphorylation of Ser(Ser-21 and Ser-9) residues (46, 48). To evaluate whetherTEAD-1 overexpression altered Akt signaling activity, we per-formed aWestern blot analysis using an anti-Akt antibody thatrecognizes all three Akt isoforms (Akt1, Akt2, andAkt3) as well

as antibodies that specifically recog-nize either Akt1 orAkt2 protein. Nodifference in total Akt (Akt1 to -3),Akt1, or Akt2 protein was detectedbetween the cellular extracts iso-lated from the EDL and plantarismuscles of WT and TEAD-1 Tgline 12 mice (Fig. 7, A and B). Simi-larly, phospho-Ser473 Akt antibodydid not detect significant differ-ences in Akt phosphorylationbetween WT and TEAD-1 Tg EDLandplantarismuscle extracts (Fig. 7,A and B). These data show that acti-vation of GSK-3� and GSK-3� inthe EDL and plantaris muscles ofTEAD-1 transgenic mice was notdue to decreases in total Akt or Aktactivity (phospho-Akt).Increased GSK-3 Activity Alters

Nuclear Levels of Its DownstreamTarget �-Catenin—It is well estab-lished that activated GSK-3� is anegative regulator of the transcrip-tional regulatory protein,�-catenin.GSK-3-mediated phosphorylationof �-catenin leads to its proteo-some-mediated degradation, result-ing in altered transcriptional regula-tion of �-catenin target genes (46,49). Immunoblot analysis usinganti-�-catenin-specificantibodyde-tected reduced levels of �-cateninin the nuclear extracts obtainedfrom the EDL and plantaris musclesof TEAD-1 Tg line 12 mice.Although we detected a decrease innuclear �-catenin, no differenceswhere obtained between WT andHA-TEAD-1 line 12 mice whenusing total cellular extracts, whichis probably due to the largeamount of total cellular �-catenin(Fig. 8, A–D).

The Effects of Activated GSK-3 Are Reversed by MOV andTEAD-1 siRNA—Previously, we have demonstrated that 48 h ofMOVwas sufficient to nearly silence the expression of a 3300-bpmouseMCKpromoter-reporter transgene (�3300to�7-CAT) inthe plantaris muscle of transgenic mice (26). Since expression ofthe HA-TEAD-1 transgene is controlled in part by a 3300-bpMCKpromoter (see Fig. 1A), we tested the hypothesis that 48 h ofMOVwould return the levels of phospho-GSK-3�/� back toWTlevels, thereby restoring nuclear NFAT and �-catenin levels.Immunoblot analysis using protein extracts isolated from 48-hMOV-plantaris muscle of TEAD-1 Tg line 12 mice and an anti-TEAD-1-specific antibody showed a remarkable decrease in HA-TEAD-1 protein when compared with plantaris muscle proteinextract obtained from TEAD-1 Tg mice that had not been sub-

FIGURE 5. Western blot analysis of NFATc1 and NFATc3 expression in fast twitch skeletal muscle. Densi-tometry quantification of nuclear protein extracts (100 �g) obtained from adult WT and TEAD-1 Tg (line 12) EDLand plantaris muscle show a significant decrease of nuclear NFAT protein in transgenic versus wild type EDL (Aand B) and plantaris (C and D) muscle. Comparative Western blot analysis using 100 �g of total protein extractobtained from either adult WT control or TEAD-1 Tg (line 12) control revealed no difference in total level ofNFAT protein in EDL (E) and plantaris (F).




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

jected toMOV (Fig. 9A).Most strikingly, 48 h ofMOV resulted ina rescue of GSK-3� and GSK-3� phosphorylation level that wasnot different from that detected in total protein extract isolatedfrom the plantaris muscle of WT and WT MOV mice (Fig. 9A,lane 1 versus lanes 4–7). The latter indicates that 48 hofMOVdidnot alter the phosphorylation status of GSK-3�/�. Furthermore,Western blot analysis using nuclear extract isolated from 48-hTEAD-1TgMOV-plantarismuscle showed that thenuclear levelsof NFATc1, NFATc3, and �-catenin were similar to thosedetected in nuclear extracts isolated from the plantaris muscle ofWTmice (Fig. 9B, lane 1 versus lanes 4–6). SinceAkt is a primaryupstream regulator of GSK-3 activity, we performed a Westernblot analysis to determine if 48 h ofMOV activated Akt signaling.The levels of total and phosphorylated Akt in TEAD-1 TgMOV-plantaris total protein extract did not differ significantly fromthose detected in plantaris extract isolated fromTEAD-1Tgmicenot subjected toMOV (Fig. 9C, lanes 1–3 versus lanes 4–6). Sim-ilar resultswereobtainedwhenassessing total andphosphorylatedAkt in total protein extract obtained fromWT versusWT-MOVplantaris muscle (Fig. 9D, lanes 1–3 versus lanes 4–6).Mouse skeletal muscle satellite cells reportedly appear

shortly before the end of gestation (2, 50, 51). Since we havedetected HA-TEAD-1 protein in the hind limb skeletal muscle

of embryonic day 17 (fetal) TEAD-1Tg line 12 mice (Fig. 1H), it wasimportant to examine if theHA-TEAD-1 transgene was ex-pressed in adult skeletal musclemyoblasts (satellite cells) of thesemice and, if so, whether TEAD-1overexpression in these cells re-sulted in an altered phosphorylationstatus of GSK-3�/�. Western blotanalysis using TEAD-1-specificantibody and total protein extractobtained from differentiated satel-lite cells isolated from either WT orTEAD-1 Tg mice (3 months old)revealed the presence of endoge-nous TEAD-1 (Fig. 9E, lanes 1–4),whereas the HA-tagged TEAD-1protein only appeared in the proteinextracts of differentiated satellitecells isolated from TEAD-1 Tg line12mice (Fig. 9E, lanes 1 and 2 versuslane 4). Transfection of TEAD-1 Tgsatellite cells with TEAD-1-specificsiRNA markedly decreased the lev-els of total cellular TEAD-1 (endog-enous and HA-TEAD-1) protein ascompared with transfection withnontargeting siRNA (Fig. 9E, lane 2versus lane 3) or in comparison withWT TEAD-1 levels (Fig. 9E, lane 3versus lane 4). Importantly, West-ern blot analysis revealed that thephosphorylation status of GSK-3�/� was strikingly decreased in

protein extracts isolated from TEAD-1 Tg satellite cells whencompared with protein extracts isolated fromWT satellite cells(Fig. 9E, lanes 1 and 2 versus lane 4). GSK-3�/� phosphoryla-tion levels returned to WT levels in TEAD-1 Tg satellite cellstreated with TEAD-1 siRNA (Fig. 9E, lane 3 versus 4), whereasthose treated with nontargeting siRNA maintained reducedlevels of phospho-GSK-3�/� (Fig. 9E, lane 2 versus lane 3).When considered collectively, these data provide convincingevidence that TEAD-1 participates in slow muscle contractileprotein gene expression in vivo and reveal a regulatory role forTEAD-1 in GSK-3�/� signaling.

DISCUSSION

This study provides the first in vivo evidence indicating thatTEAD-1 participates in slow skeletal muscle gene expression,and it reveals that TEAD-1 unexpectedly activated GSK-3�/�,resulting in decreased nuclear �-catenin and NFATc1/c3 pro-tein. We have verified the link between increased TEAD-1expression and activation of GSK-3�/� by using TEAD-1siRNA in culture satellite cells and in vivo by using MOV, bothof which decreased TEAD-1 transgene expression, therebyrestoringWT levels of phospho-GSK-3�/� andnuclear�-cateninandNFATc1/c3.TEAD-1activationofGSK-3�/� signaling in sat-

FIGURE 6. Western blot analysis of total and phosphorylated GSK-3� and -� expression in fast twitchskeletal muscle. Densitometry quantification of total protein extracts (100 �g) obtained from adult WT andTEAD-1 Tg (line 12) EDL and plantaris muscle reveals no difference in total GSK-3� and GSK-3� protein and asignificant decrease in the phosphorylation status of GSK-3� and GSK-3� in transgenic versus wild type EDL (A)and plantaris (B) muscle.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

ellite cells may have future applica-tions tonovel therapeutic strategies inregenerative medicine.Although numerous in vitro stud-

ies have provided evidence thatTEAD proteins regulate basal andadaptive skeletal muscle contractileprotein gene expression, their invivo role in this capacity has notbeen examined in adult skeletalmuscle. We show herein that a per-sistent increase in TEAD-1 expres-sion in adult skeletal muscleinduced a transition toward a slowmuscle contractile protein pheno-type. This was evidenced by a shifttoward a slower MyHC profile thatwas distinguished by the virtual lossof fast type IIx/d MyHC protein inboth fast and slow twitch muscles, avariable decrease in type IIb MyHCprotein, and an increase in type IIaand/or type I MyHC protein foreach of three independent Tg lines(Fig. 3E). Most importantly, thephysiological significance of theshift toward a slowerMyHC expres-sion pattern was underscored by acorresponding decrease in theshortening velocity of the fast twitchEDL muscle, which is consistentwith a large body of evidence dem-onstrating that each MyHC isopro-tein confers a different ATPaseactivity and contractile velocity (3,41, 52). In addition to the transitiontoward a slower MyHC expressionpattern in the fast twitch EDLmuscle, we have shown a strikingincrease in the abundance ofmRNAs encoding the slow isoformsof the thin filament troponin com-plex proteins (TnC, TnT, and TnI(troponin C, troponin T, and tro-ponin I, respectively)) and the slowsarcoplasmic calcium ATPase(SERCA2a) that was paralleled byincreases in myoglobin, slow TnI,and SERCA2a protein with adecrease in fast SERCA1mRNAandprotein. This fast-to-slow transitionin EDL gene expression was paral-leled by significantly longer con-traction and relaxation times, whichcorresponds well with previousstudies demonstrating that the tro-ponin complex proteins and sarco-plasmic calcium-ATPase regulate

FIGURE 7. Western blot analysis of total and phosphorylated Akt protein in fast-twitch skeletal muscle.Densitometry quantification of total protein extracts (100 �g) obtained from WT and TEAD-1 Tg (line 12) EDLand plantaris muscle revealed no difference in expression and phosphorylation levels of Akt in transgenic micein EDL (A) and plantaris (B) muscle.

FIGURE 8. Western blot analysis of nuclear �-catenin expression in fast-twitch skeletal muscle. Densitom-etry quantification of nuclear protein extracts (100 �g) obtained from WT and TEAD-1 Tg (line 12) EDL andplantaris muscle revealed a significant decrease in nuclear �-catenin protein abundance in transgenic EDL (A)and plantaris (B) muscle. Comparative Western blot analysis using 100 �g of total protein extract obtained fromeither adult WT control or TEAD-1 Tg (line 12) control revealed no difference in total level of �-catenin proteinin EDL (C) and plantaris (D).




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

the kinetics of calcium-activated muscle contraction (29). It isnotable that the skeletal muscle phenotype displayed by theTEAD-1 Tg mice shows a remarkable correlation to our previ-ous work wherein we showed that MOV not only inducedTEAD-1 protein expression but also increased the expressionofmyoglobin, slow TnI, and �MyHCprotein concurrent with asignificant increase in type I fibers (15, 53, 54). In addition,trends in MyHC gene expression similar to those reportedherein have been reported to occur in both human and rodentfast twitch muscle in response to resistance and enduranceexercise (55). It is also noteworthy that the transcriptional acti-vation ofmyoglobin and slowTnI in fast twitchmuscle of trans-genic mice expressing constitutively active forms of cal-cineurin, CaMK, or PDK1 has been used as a standard toconfirm the induction of a slow gene program (5, 56, 57). Col-lectively, these data provide in vivo evidence that TEAD-1serves a role in slow and oxidative skeletal muscle contractileprotein gene expression and point toward a role for this tran-scription factor in human and rodent skeletal muscle adapta-tions in response to various modes of exercise.An interesting and rather unexpected finding from our study

was that although the fast-twitch EDL muscle of TEAD-1 Tgmice displayed increased levels of �MyHC mRNA (�14-fold;this is a muscle that normally does not express �MyHC to anyappreciable degree), increased �MyHC protein expression wasonly detected in the gastrocnemius and soleusmuscles, indicat-ing that �MyHC protein production was being regulated by amuscle-specific post-transcriptional mechanism. There aretwo reasonable explanations that can account for this finding.Micro-RNAs (small noncoding RNAs) have been shown to reg-ulate gene expression post-transcriptionally by binding targetmRNAs and in this manner prevent and/or decrease proteinsynthesis of the encoded protein (58). In this regard, severalstriatedmuscle-specific micro-RNAs have been shown to servecrucial regulatory roles during skeletal muscle developmentand in cardiac gene expression and stress responsiveness (59,60). For example, the hearts from miR-208 nullizygous micedisplayed up-regulation of numerous fast twitch skeletal mus-cle genes andwere unable to induce�MyHCgene expression inresponse to pressure overload, indicating a specific role formiR-208 in maintaining the cardiac phenotype and in �MyHCgene expression as a result of stress. Although speculative, it isconceivable that the fast twitch EDL muscle expresses a set ofmicro-RNAs that target mRNAs encoding various slowmuscleproteins (e.g. �MyHC), thereby maintaining its fast fiber phe-notype. Equally plausible, the �MyHC protein may have beendegraded as an inappropriately expressed protein. The ubiq-uitin-proteasome system participates in the regulation ofnumerous cellular processes by proteolytic degradation of reg-ulatory proteins of signaling and gene expression pathways andserves as the primary site of skeletal muscle contractile proteinmaintenance and turnover under basal conditions and in

FIGURE 9. MOV- and TEAD-1-specific siRNA repress TEAD-1 transgeneexpression and reverse the effects of activated GSK-3�/� on nuclearabundance of NFATc1/c3 and �-catenin. A–D, comparative Western blotanalysis using 100 �g of total protein extract obtained from either adult WTand MOV-treated (WT MOV) or TEAD-1 Tg line 12 (Tg, L12) control and MOV-treated plantaris muscles (Tg, L12, MOV) revealed that the phosphorylation of

GSK-3�/� returned to WT levels, thereby restoring nuclear NFAT and �-cate-nin levels, whereas Akt was not activated. E, Western blot analysis of proteinextract obtained from satellite cells isolated from either WT or TEAD-1 Tg (line12) muscle treated with nontargeting or TEAD-1 siRNA showed silencing ofHA-TEAD-1 expression concurrent with a return of GSK-3�/� phosphoryla-tion status to WT levels.




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

response to physiological and pathological stimuli (61). In thisregard, mice deficient for two E3 ligases, muscle RING finger 1and 3 (MuRF1 and MuRF3), develop skeletal muscle patholo-gies characterized by subsarcolemma accumulation of MyHC(�MyHC and fast type IIa MyHC), indicating a role for MuRF1and MuRF3 in the maintenance of basal contractile proteinturnover (62). Thus, degradation of the �MyHC protein mayhave served as a compensatory response to retain the fast fiberphenotype of the EDL muscle. Clearly, skeletal muscle fibertype conversion, be it fast-to-slow or slow-to-fast, requirescoordinate regulation at multiple levels, given that it involves amultitude of changes in muscle-specific gene expression, andregulation can be imposed at any point along the gene expres-sion pathway. Consistent with this concept, recent inquiriesinto the molecular mechanisms involved in establishing, main-taining, andmodulating skeletalmuscle fiber types have led to aproposed model wherein the cooperative interactions betweenthree signaling pathways (calcineurin, CaMKs, and PKD1)modulate the transcriptional activation of slow muscle fibergenes through class II HDAC degradation and the activation ofmembers of the MEF2 and NFAT transcription factor families(5–7, 43).Another intriguing and unexpected finding was that the

TEAD-1-induced transition toward a slowermuscle contractilephenotype was accomplished in a seemingly NFAT-independ-ent manner. Our Western blot analyses showed a reduction inthe phosphorylation status of both isoforms of glycogen syn-thase kinase-3 (GSK-3� and GSK-3�; dephosphorylation acti-vates GSK kinase activity) concurrent with decreased nuclearlevels of NFATc1 and NFATc3. The observed decrease in thephosphorylation status of GSK-3�/� could not be attributed toan alteration in signaling fromAkt, a primary upstream regula-tor of GSK phosphorylation, since we did not detect changes intotal or phosphorylated levels of Akt. Importantly, previouswork has shown that GSK-3� works in opposition to the pro-tein phosphatase calcineurin. Once activated by a sustainedincrease in intracellular calcium, calcineurin dephosphorylatescytoplasmic NFAT, resulting in its nuclear translocation andsubsequent activation of NFAT-dependent gene expression,whereas GSK-3� phosphorylates NFAT, affecting its nuclearabundance and, by this means, decreases NFAT target geneexpression (63). Consistent with our result showing that GSK-3�/� can regulate nuclear NFAT levels, it has been shown thatoverexpression of a constitutively active form of GSK-3�increased NFAT export from the nucleus of cardiomyocytes(64). Furthermore, the inhibition of GSK-3� has been shown toslow the export of NFATc1 from the nucleus of single fibersisolated from themouse flexor digitorum brevis muscle follow-ing the cessation of electrically stimulated import of NFATc1(65). It is also worth noting that although several studies haveprovided evidence that calcineurin signaling is important forfiber type switching, activation of the slow gene program canoccur by an NFAT-independent mechanism (13, 56, 57,66–69).AswithNFAT,we also demonstrate a notable decrease in the

nuclear levels of�-catenin, which is another downstream targetof activated GSK-3�. This result is consistent with previouswork, wherein a decrease in �-catenin was detected in embry-

onic stem cells isolated from double homozygous GSK�/�(21A/21A/9A/9A) knock-in mice expressing constitutively activeforms of GSK-3�/� due to a serine-to-alanine substitution thatprevents phosphorylation and thus inactivation of GSK-3� andGSK-3� (70). GSK-3� phosphorylation of �-catenin targets itfor degradation by the ubiquitin proteosome, thereby control-ling its nuclear level and transcriptional activity of �-catenin/TCF/Lef-responsive promoters (46, 47). Although �-cateninhas been shown to be necessary for skeletal muscle develop-ment (71), we did not detect a difference in skeletal musclegrowth rate or normalized weight between wild type andTEAD-1 Tg mice despite the early expression pattern of theTEAD-1 transgene (embryonic day 17), indicating that �-cate-nin is required at an earlier time during myogenesis or that theoverall magnitude of decrease in nuclear �-catenin in ourexperiment was not sufficient to altermuscle growth. Althoughour data did not provide evidence as to whether decreasednuclear �-catenin contributed to the observed transition to aslower contractile protein phenotype, a role in this capacity isconceivable based on the numerous TCF/Lef elements con-tained within the promoter regions of the MyHC, troponincomplex, and SERCA2a genes.Our current study also provides substantial evidence that the

persistent increase in TEAD-1 expression is responsible for thereduced level of phospho-GSK-3�/3� and nuclear NFATc1/3and �-catenin. This was evidence by employing 48 h of MOV,which resulted in a significant decrease in MCK-driven HA-tagged TEAD-1 transgene expression and a return of phospho-GSK-3�/3�, nuclear NFATc1/c3, and �-catenin to WT levelsin the absence of a significant change in total or phospho-Akt.Of direct relevance to the latter study is our previous workshowing that 48 h of MOV is sufficient to repress endogenousand transgenic MCK reporter gene expression (26). However,sinceMOV is associated with an increase in motor nerve activ-ity, a stimulus previously shown to regulate nucleocytoplasmicshuttling of NFAT via the calcium-dependent calcineurin-NFATpathway (65, 72, 73), it could be argued that the observedreturn of nuclear NFATc1/c3 to wild type levels in our studywas due to heightened neural activity imposed by MOV. Nev-ertheless, the connection between increased TEAD-1 expres-sion and dephosphorylation of GSK-3�/� was further substan-tiated when we demonstrated that adult stage myoblast(satellite cells) isolated from TEAD-1 Tg mice displayed WTlevels of phospho-GSK-3�/� concurrent with a strikingdecrease in MCK-driven HA-TEAD-1 expression as a result oftransfection with TEAD-1-specific siRNA. Although beyondthe scope of the current study, it will be intriguing to determinethe role TEAD-1 serves in satellite cell (skeletal muscle adultstem cells) regulation, given that satellite cells contribute topostnatal skeletal muscle growth and play an essential role inskeletal muscle homeostasis and regeneration after injury andduring aging (74).In summary, the present study provides the first in vivo evi-

dence that TEAD-1 participates in slow skeletal muscle geneexpression and revealed a novel link between increasedTEAD-1 expression and activation of GSK-3�/�, which, inturn, decreased nuclear NFATc1/c3 and�-catenin protein. It isexpected that this putative pathway (TEAD-13GSK-3�/�3




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

�-catenin/NFATc1/c3) would work in conjunction with thewell established calcium-dependent signaling pathways to acti-vate a complete array of gene programs necessary to accom-plish a complete fast-to-slow phenotype switch in adult skeletalmuscle. However, our data also support the notion thatTEAD-1 can induce a transition toward a slower contractileprotein phenotype independent of activation of calcium-de-pendent pathways, conceivably via direct TEAD element bind-ing and combinatorial interactions with adjacently bound tran-scriptional regulators and coactivators. Additional work will berequired to elucidate 1) the phosphatase that dephosphorylatesGSK-3�/�, 2) the transcription factors and coactivators thatinteract with the TEAD proteins in vivo to mediate skeletalmuscle-fiber-specific gene expression under basal conditionsand during remodeling in response to MOV/exercise, and 3)the full complement of TEAD target genes, given the crucialtranscriptional role recently reported for the TEAD proteins inneural cells and smooth and striated muscle (17–22, 75).Finally, our current results suggest a possible regulatory role ofTEAD-1 in satellite cells. Because satellite cells are basicallysolely responsible for regeneration of adult skeletal muscle,increased TEAD-1 expression potentially may be exploited forthe development of novel therapeutic strategies in regenerativemedicine.

Acknowledgments—We thankDrs. DawnCornelison andMarkHan-nink for critical review of the manuscript and Dr. Fadia Haddad forcritical advice on MyHC gel separation. The transgenic mice wereproduced by the University of Missouri-Columbia TransgenicFacility.

REFERENCES1. Schiaffino, S., and Reggiani, C. (1996) Physiol. Rev. 76, 371–4232. Biressi, S., Molinaro, M., and Cossu, G. (2007) Dev. Biol. 308, 281–2933. Barany, M. (1967) J. Gen. Physiol. 50, (suppl.) 197–2184. Booth, F.W., and Baldwin, K.M. (1996) inHandbook of Physiology, Section

12: Exercise: Regulation and Integration of Multiple Systems (Rowell, L. B.,and Sheperd, J. T., eds) pp. 1075–1123, American Physiology Society,Bethesda, MD

5. Kim,M. S., Fielitz, J., McAnally, J., Shelton, J.M., Lemon, D. D.,McKinsey,T. A., Richardson, J. A., Bassel-Duby, R., andOlson, E. N. (2008)Mol. Cell.Biol. 28, 3600–3609

6. Potthoff, M. J., Olson, E. N., and Bassel-Duby, R. (2007)Curr. Opin. Rheu-matol. 19, 542–549

7. Potthoff, M. J., Wu, H., Arnold, M. A., Shelton, J. M., Backs, J., McAnally,J., Richardson, J. A., Bassel-Duby, R., andOlson, E. N. (2007) J. Clin. Invest.117, 2459–2467

8. Grifone, R., Laclef, C., Spitz, F., Lopez, S., Demignon, J., Guidotti, J. E.,Kawakami, K., Xu, P. X., Kelly, R., Petrof, B. J., Daegelen, D., Concordet,J. P., and Maire, P. (2004)Mol. Cell. Biol. 24, 6253–6267

9. Issa, L. L., Palmer, S. J., Guven, K. L., Santucci, N., Hodgson, V. R., Popovic,K., Joya, J. E., and Hardeman, E. C. (2006) Dev. Biol. 293, 104–115

10. Arany, Z., Lebrasseur, N., Morris, C., Smith, E., Yang,W., Ma, Y., Chin, S.,and Spiegelman, B. M. (2007) Cell Metab. 5, 35–46

11. Schuler, M., Ali, F., Chambon, C., Duteil, D., Bornert, J. M., Tardivel, A.,Desvergne, B., Wahli, W., Chambon, P., and Metzger, D. (2006) CellMetab. 4, 407–414

12. Barish, G. D., Narkar, V. A., and Evans, R. M. (2006) J. Clin. Invest. 116,590–597

13. Calvo, S., Vullhorst, D., Venepally, P., Cheng, J., Karavanova, I., and Buo-nanno, A. (2001)Mol. Cell. Biol. 21, 8490–8503

14. Vyas, D. R., McCarthy, J. J., and Tsika, R. W. (1999) J. Biol. Chem. 274,

30832–3084215. Karasseva, N., Tsika, G., Ji, J., Zhang, A.,Mao, X., and Tsika, R. (2003)Mol.

Cell. Biol. 23, 5143–516416. Larkin, S. B., andOrdahl, C. P. (1999) inHeart Development (Harvey, R. P.,

and Rosenthal, N., eds) pp. 307–329, Academic Press, Inc., San Diego17. Yoshida, T. (2008) Arterioscler. Thromb. Vasc. Biol. 28, 8–1718. Sawada, A., Kiyonari, H., Ukita, K., Nishioka, N., Imuta, Y., and Sasaki, H.

(2008)Mol. Cell. Biol., in press19. Kaneko, K. J., Kohn, M. J., Liu, C., and DePamphilis, M. L. (2007) Genesis

45, 577–58720. Nishioka, N., Yamamoto, S., Kiyonari, H., Sato, H., Sawada, A., Ota, M.,

Nakao, K., and Sasaki, H. (2008)Mech. Dev. 125, 270–28321. Yagi, R., Kohn, M. J., Karavanova, I., Kaneko, K. J., Vullhorst, D., DePam-

philis, M. L., and Buonanno, A. (2007) Development 134, 3827–383622. Chen, Z., Friedrich, G. A., and Soriano, P. (1994)Genes Dev. 8, 2293–230123. Johnson, J. E., Wold, B. J., and Hauschka, S. D. (1989) Mol. Cell. Biol. 9,

3393–339924. Nguyen, Q. G., Buskin, J. N., Himeda, C. L., Fabre-Suver, C., and Haus-

chka, S. D. (2003) Transgenic Res. 12, 337–34925. Nguyen, Q. G., Buskin, J. N., Himeda, C. L., Shield, M. A., and Hauschka,

S. D. (2003) J. Biol. Chem. 278, 46494–4650526. Tsika, R. W., Hauschka, S. D., and Gao, L. (1995) Am. J. Physiol. 269,

C665–C67427. Tsika, R. W. (1994) Exercise Sport Sci. Rev. 22, 361–38828. Bruning, J. C., Michael, M. D., Winnay, J. N., Hayashi, T., Horsch, D.,

Accili, D., Goodyear, L. J., and Kahn, C. R. (1998)Mol. Cell 2, 559–56929. Sonnemann, K. J., Fitzsimons, D. P., Patel, J. R., Liu, Y., Schneider Martin,

F., Moss, R. L., and Ervasti, James, M. (2006) Dev. Cell 11, 387–39730. Lynch, G. S., Hinkle, R. T., Chamberlain, J. S., Brooks, S. V., and Faulkner,

J. A. (2001) J. Physiol. (Lond.) 535, 591–60031. Herron, T. J., Korte, F. S., and McDonald, K. S. (2001) Am. J. Physiol. 281,

H1217–H122232. Talmadge, R. J., and Roy, R. R. (1993) J. Appl. Physiol. 75, 2337–234033. Cornelison, D. D.W.,Wilcox-Adelman, S. A., Goetinck, P. F., Rauvala, H.,

Rapraeger, A. C., and Olwin, B. B. (2004) Genes Dev. 18, 2231–223634. Trask, R. V., and Billadello, J. J. (1990) Biochim. Biophys. Acta 1049,

182–18835. Lyons, G. E., Muhlebach, S., Moser, A., Masood, R., Paterson, B.M., Buck-

ingham, M. E., and Perriard, J. C. (1991) Development 113, 1017–102936. Vyas, D. R., McCarthy, J. J., Tsika, G. L., and Tsika, R. W. (2001) J. Biol.

Chem. 276, 1173–118437. Larkin, S. B., Farrance, I. K., and Ordahl, C. P. (1996) Mol. Cell. Biol. 16,

3742–375538. Butler, A. J., and Ordahl, C. P. (1999)Mol. Cell. Biol. 19, 296–30639. Stewart, A. F., Larkin, S. B., Farrance, I. K., Mar, J. H., Hall, D. E., and

Ordahl, C. P. (1994) J. Biol. Chem. 269, 3147–315040. Farrance, I. K., and Ordahl, C. P. (1996) J. Biol. Chem. 271, 8266–827441. Bottinelli, R., Betto, R., Schiaffino, S., and Reggiani, C. (1994) J. Physiol.

478, 341–34942. Moss, R. L., Diffee, G.M., and Greaser, M. L. (1995) Rev. Physiol. Biochem.

Pharmacol. 126, 1–6343. Bassel-Duby, R., and Olson, E. N. (2006) Annu. Rev. Biochem. 75, 19–3744. Beals, C. R., Sheridan, C.M., Turck, C.W., Gardner, P., andCrabtree, G. R.

(1997) Science 275, 1930–193445. Neal, J. W., and Clipstone, N. A. (2001) J. Biol. Chem. 276, 3666–367346. Doble, B. W., and Woodgett, J. R. (2003) J. Cell Sci. 116, 1175–118647. Forde, J. E., and Dale, T. C. (2007) Cell Mol. Life Sci. 64, 1930–194448. Manning, B. D., and Cantley, L. C. (2007) Cell 129, 1261–127449. Eastman, Q., and Grosschedl, R. (1999)Curr. Opin. Cell Biol. 11, 233–24050. Relaix, F., Rocancourt, D., Mansouri, A., and Buckingham, M. (2005) Na-

ture 435, 948–95351. Hawke, T. J., and Garry, D. J. (2001) J. Appl. Physiol. 91, 534–55152. Bottinelli, R., Schiaffino, S., and Reggiani, C. (1991) J. Physiol. 437,

655–67253. Tsika, G. L., Wiedenman, J. L., Gao, L., McCarthy, J. J., Sheriff-Carter, K.,

Rivera-Rivera, I. D., and Tsika, R. W. (1996) Am. J. Physiol. 271,C690–C699

54. Wiedenman, J. L., Rivera-Rivera, I., Vyas, D., Tsika, G., Gao, L., Sheriff-




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Carter, K., Wang, X., Kwan, L. Y., and Tsika, R. W. (1996) Am. J. Physiol.270, C1111–C1121

55. Baldwin, K. M., and Haddad, F. (2001) J. Appl. Physiol. 90, 345–35756. Wu, H., Naya, F. J., McKinsey, T. A., Mercer, B., Shelton, J. M., Chin, E. R.,

Simard, A. R., Michel, R. N., Bassel-Duby, R., Olson, E. N., and Williams,R. S. (2000) EMBO J. 19, 1963–1973

57. Wu, H., Rothermel, B., Kanatous, S., Rosenberg, P., Naya, F. J., Shelton,J. M., Hutcheson, K. A., DiMaio, J. M., Olson, E. N., Bassel-Duby, R., andWilliams, R. S. (2001) EMBO J. 20, 6414–6423

58. Krutzfeldt, J., and Stoffel, M. (2006) Cell Metab. 4, 9–1259. Kim, H. K., Lee, Y. S., Sivaprasad, U., Malhotra, A., and Dutta, A. (2006)

J. Cell Biol. 174, 677–68760. van Rooij, E., and Olson, E. N. (2007) J. Clin. Invest. 117, 2369–237661. Naujokat, C., and Saric, T. (2007) Stem Cells 25, 2408–241862. Fielitz, J., Kim, M. S., Shelton, J. M., Latif, S., Spencer, J. A., Glass, D. J.,

Richardson, J. A., Bassel-Duby, R., and Olson, E. N. (2007) J. Clin. Invest.117, 2486–2495

63. Crabtree, G. R., and Olson, E. N. (2002) Cell 109, (suppl.) 67–7964. Haq, S., Choukroun, G., Kang, Z. B., Ranu, H., Matsui, T., Rosenzweig, A.,

Molkentin, J. D., Alessandrini, A.,Woodgett, J., Hajjar, R.,Michael, A., andForce, T. (2000) J. Cell Biol. 151, 117–130

65. Shen, T., Cseresnyes, Z., Liu, Y., Randall, W. R., and Schneider, M. F.(2007) J. Physiol. (Lond.) 579, 535–551

66. Delling, U., Tureckova, J., Lim, H. W., De Windt, L. J., Rotwein, P., andMolkentin, J. D. (2000)Mol. Cell. Biol. 20, 6600–6611

67. Naya, F. J., Mercer, B., Shelton, J., Richardson, J. A., Williams, R. S., andOlson, E. N. (2000) J. Biol. Chem. 275, 4545–4548

68. Swoap, S. J. (1998) Am. J. Physiol. 274, C681–C68769. Parsons, S. A.,Wilkins, B. J., Bueno, O. F., andMolkentin, J. D. (2003)Mol.

Cell. Biol. 23, 4331–434370. McManus, E. J., Sakamoto, K., Armit, L. J., Ronaldson, L., Shpiro, N.,

Marquez, R., and Alessi, D. R. (2005) EMBO J. 24, 1571–158371. Cossu, G., and Borello, U. (1999) EMBO J. 18, 6867–687272. Tothova, J., Blaauw, B., Pallafacchina, G., Rudolf, R., Argentini, C., Reggi-

ani, C., and Schiaffino, S. (2006) J. Cell Sci. 119, 1604–161173. Schiaffino, S., Sandri, M., and Murgia, M. (2007) Physiol. (Bethesda) 22,

269–27874. Kuang, S., Gillespie, M. A., and Rudnicki, M. A. (2008) Cell Stem Cell 2,

22–3175. Gan, Q., Yoshida, T., Li, J., and Owens, G. K. (2007) Circ. Res. 101,

883–892




ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

L. Moss and Juan JiRichard W. Tsika, Christine Schramm, Gretchen Simmer, Daniel P. Fitzsimons, Richard

Slower Skeletal Muscle Contractile PhenotypeOverexpression of TEAD-1 in Transgenic Mouse Striated Muscles Produces a

doi: 10.1074/jbc.M807461200 originally published online October 31, 20082008, 283:36154-36167.J. Biol. Chem.

10.1074/jbc.M807461200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/283/52/36154.full.html#ref-list-1

This article cites 72 references, 32 of which can be accessed free at


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M807461200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;283/52/36154&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/283/52/36154

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=283/52/36154&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/283/52/36154

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/283/52/36154.full.html#ref-list-1

http://www.jbc.org/

overexpressionoftead-1intransgenicmousestriatedmuscles...

Documents